Julia - Download Dataframes from the Internet

Installation

For a download to a CSV format, you need to install the package CSV. In the package manager, add the package “CSV”.

add CSV

At the top of your program, add

using DataFrames
using CSV

First example: Some Programming Languages

At first, you make the request to get the content from a specified URL.

url = "https://raw.githubusercontent.com/nassarhuda/easy_data/master/programming_languages.csv"
D = download(url, "programming_languages.csv")

The CSV functionality converted the data to a dataframe.

Save the csv-file to the local file system

Specify a name:

filename = "programming_languages.csv"

If you want to use the name of this current file, you can get it from

@__FILE__
CSV.write(filename, df)
println("\nFile written to $filename")

If you want to use the current path, you can get it from

absPath = @__DIR__
absPath = pwd()

Example: Some CO2 - concentration

Just to make another example. This time using a relative path for saving the downloaded data.

path = "./ZZ_Data/"
filename = "co2.csv"
relname = path * filename

The proceed as usual:

url2 ="https://raw.githubusercontent.com/vega/vega-datasets/master/data/co2-concentration.csv"
D = download(url2, relname)
Copyright 2021-2022